[Robotgroup] Detect the Rink. Re: Drive a Droid proposal

[brooksdesign]

One other way of doing this may be like my BlockHeads robots where the objective IS to knock your opponent out of the ring and a roboteer standing by just resets them to their starting corners. By the way I have been cleaning them up after posting the new video http://wiki.therobotgroup.org/wiki/BlockHeads and was thinking of donating them to the group for use in outreach events, especialy if someone wanted to convert them to RC and cut the cords.
-brooks

-----Original Message-----
>From: "john at cozmicfunk.com" <john at cozmicfunk.com>
>Sent: Jul 8, 2008 2:09 AM
>To: The Robot Group Mailing List <robotgroup at puremagic.com>
>Subject: Re: [Robotgroup] Detect the Rink.  Re: Drive a Droid proposal
>
>My personal opinion: I think you are going about this design the wrong 
>way...
>
>Why does it have to have unlimited RC control then only be able to stay 
>in a small circle? it makes no sense at all...
>
>Solution: You need to make the Robot control idiot proof! Instead of 
>giving COMPLETE rc control, you should design the control features 
>around a few pre-programed presets that allow some function (with a 
>avoidance detection system) and also give the audience what they want: 
>Interaction!
>
>Because from my point of view, I have already been cruising around a 
>small RC bot and it gets old really fast.
>
>So the idea is have some presets like the smaller voice controlled r2d2 
>(which I have) and design it to stay within a perimeter of function and 
>then return to a Center location or Homing beacon of some kind after it 
>does it's two minute Parade. It can still be entertaining and engaging 
>without it being this huge design concept that will only work with more 
>money and resources than it will be worth for only a First night activity.
>
>The small Voice activated R2 has a neat feature called Dance mode where 
>it drives around playing the Cantina theme and beeps and lights up. for 
>kids, it is great, they don't have to be in control to enjoy it and then 
>there is no liability with it going crazy and running someone over either.
>
>Like I said, you could have a few presets and some demo modes to mix it 
>up, maybe have a ask R2 voice system that randomly picks from a  list of 
>preset beeping patterns that would appear to be ":responsive" and no one 
>would ever know the difference.
>
>Take this advice from a guy that has been designing a sci-fi film for 
>years and has pondered ways of making things easy, simple and effective 
>in communicating the ideas to an audience. People only know what ever 
>you tell them: give them some interaction with lights and sounds and 
>they will be happy :)
>
>JPF
>
>
>
>Bruce Waters wrote:
>> Vern,
>>         I suggest a point-of-view shift from detect-the-ropes
>> to detect-the-rink.   If you place two detectors, one outboard
>> of each wheel,  you can have the bot naturally turn back into 
>> the rink without a rink monitor person.   Just "pause" the wheel 
>> *opposite* the sensor which no longer senses the rink.
>> As the "guest operator" continues forward, the bot will curve 
>> back into the rink.   Even if the "guest operator" goes 
>> backwards, the paused wheel (pls recall: that is the one probably 
>> still on the rink) will just hold its ground as the bot swivels the 
>> out-of-bounds wheel back onto the rink.  Also, if neither 
>> wheel senses the rink then both wheels are "paused" and the 
>> way-out-of-bounds bot simply stops.  An override to take a
>> bot out of "rink only mode" could be available for a rink 
>> monitor administrative person if desired or one could use the
>> brute-force technique of picking up a way-out-of bounds bot 
>> and placing it back on the rink.
>>       A down side to the detect-the-rink approach is that the 
>> entire rink area (less swivel space) has to be made so the 
>> sensors can detect the rink and still do not false-positive 
>> when a bot is out-of-bounds.    
>>    False negatives (from perhaps a failing sensor) would show
>> up as a bot on the rink going in tight circles.   With some 
>> "handed" sensors one might even be able to globally trigger 
>> left-or-right false negatives and have all the bots act like 
>> ballroom dancers synchronously twirling CW or CCW to 
>> delight any bot choreographer or racetrack anarchist.
>>      A great advantage of detect-the-rink is that the "rink" can 
>> be any shape including shapes with safety zones.   Some 
>> interesting shapes might include a "racetrack figure8" or oval
>> wide enough for bots to pass each other.  Ropes for crowd 
>> control are orthogonal to robot control with detect-the-rink .  
>> One could choose to have an inner and an outer oval of ropes 
>> for a racetrack oval, or just an outer set, or no ropes at all.   
>> One could even have segmented ropes with gates or holes 
>> and the ropes could move around off-rink without interfering 
>> with bot control.  
>>      
>> Bruce Waters
>>      
>> _______________________________________________
>> Robotgroup mailing list
>> Robotgroup at puremagic.com
>> http://lists.puremagic.com/cgi-bin/mailman/listinfo/robotgroup
>>
>> No virus found in this incoming message.
>> Checked by AVG - http://www.avg.com 
>> Version: 8.0.138 / Virus Database: 270.4.6/1539 - Release Date: 7/7/2008 6:35 PM
>>
>>
>>
>>   
>
>_______________________________________________
>Robotgroup mailing list
>Robotgroup at puremagic.com
>http://lists.puremagic.com/cgi-bin/mailman/listinfo/robotgroup


________________________________________
PeoplePC Online
A better way to Internet
http://www.peoplepc.com